Update app.py

app.py

@@ -8,7 +8,6 @@ import edge_tts
 import streamlit.components.v1 as components
 
 # -------------------- Configuration --------------------
-# Exactly 11 user names and 11 voices (as an example)
 USER_NAMES = [
     "Aria", "Guy", "Sonia", "Tony", "Jenny", "Davis", "Libby", "Clara", "Liam", "Natasha", "William"
 ]
@@ -20,24 +19,18 @@ ENGLISH_VOICES = [
 ]
 
 USER_VOICES = dict(zip(USER_NAMES, ENGLISH_VOICES))
-
 SAVED_INPUTS_DIR = "saved_inputs"
 os.makedirs(SAVED_INPUTS_DIR, exist_ok=True)
 
-# Session state
 if 'user_name' not in st.session_state:
     st.session_state['user_name'] = USER_NAMES[0]
-
 if 'old_val' not in st.session_state:
     st.session_state['old_val'] = None
-
 if 'should_rerun' not in st.session_state:
     st.session_state['should_rerun'] = False
-
 if 'viewing_prefix' not in st.session_state:
     st.session_state['viewing_prefix'] = None
 
-# -------------------- Utility Functions --------------------
 def clean_for_speech(text: str) -> str:
     text = text.replace("\n", " ")
     text = text.replace("</s>", " ")
@@ -118,16 +111,26 @@ def arxiv_search(query, max_results=3):
         for entry in entries:
             title = entry.find('a:title', ns).text.strip()
             summary = entry.find('a:summary', ns).text.strip()
+            # Extract links (PDF) if available
+            links = entry.findall('a:link', ns)
+            pdf_link = None
+            for link in links:
+                if link.get('type') == 'application/pdf':
+                    pdf_link = link.get('href')
             summary_short = summary[:300] + "..."
-            results.append((title, summary_short))
+            # Include PDF link and title
+            if pdf_link:
+                formatted = f"Title: {title}\nPDF: {pdf_link}\nSummary: {summary_short}"
+            else:
+                formatted = f"Title: {title}\n(No PDF link)\nSummary: {summary_short}"
+            results.append(formatted)
         return results
     return []
 
 def summarize_arxiv_results(results):
-    lines = []
-    for i, (title, summary) in enumerate(results, 1):
-        lines.append(f"Result {i}: {title}\n{summary}\n")
-    return "\n\n".join(lines)
+    if not results:
+        return "No results found."
+    return "\n\n".join(results)
 
 def concatenate_mp3(files, output_file):
     with open(output_file, 'wb') as outfile:
@@ -135,21 +138,6 @@ def concatenate_mp3(files, output_file):
             with open(f, 'rb') as infile:
                 outfile.write(infile.read())
 
-def load_groups():
-    files = list_saved_inputs()
-    groups = defaultdict(list)
-    for fpath in files:
-        fname = os.path.basename(fpath)
-        prefix = fname[:10]
-        groups[prefix].append(fpath)
-    for prefix in groups:
-        groups[prefix].sort(key=lambda x: os.path.getmtime(x), reverse=True)
-    sorted_prefixes = sorted(groups.keys(),
-                             key=lambda pre: max(os.path.getmtime(x) for x in groups[pre]),
-                             reverse=True)
-    return groups, sorted_prefixes
-
-# -------------------- Main Application --------------------
 st.title("🎙️ Voice Chat & ArXiv Search")
 
 with st.sidebar:
@@ -169,55 +157,45 @@ with st.sidebar:
         st.success("All history cleared!")
         st.experimental_rerun()
 
-# Voice input component (replace path with your component)
 mycomponent = components.declare_component("mycomponent", path="mycomponent")
 voice_val = mycomponent(my_input_value="Start speaking...")
 
-tabs = st.tabs(["🎤 Voice Chat", "🔍 ArXiv Search", "💾 History", "⚙️ Settings"])
+tabs = st.tabs(["🎤 Voice Chat", "💾 History", "⚙️ Settings"])
 
 # ------------------ Voice Chat Tab -------------------------
 with tabs[0]:
     st.subheader("🎤 Voice Chat")
     if voice_val:
         voice_text = voice_val.strip()
-        edited_input = st.text_area("✏️ Edit Voice Input:", value=voice_text, height=100)
-        autorun = st.checkbox("⚡ Auto-Run", value=True)
         input_changed = (voice_text != st.session_state.get('old_val'))
+        if input_changed and voice_text:
+            # 1. Save user input
+            save_input_as_md(st.session_state['user_name'], voice_text, prefix="input")
 
-        if autorun and input_changed:
-            st.session_state['old_val'] = voice_text
-            # Save input right away
-            saved_path = save_input_as_md(st.session_state['user_name'], edited_input, prefix="input")
-            st.success("Saved input!")
-
-        if st.button("📝 Save Input Manually"):
-            saved_path = save_input_as_md(st.session_state['user_name'], edited_input, prefix="input")
-            st.success("Saved input!")
-
-    st.write("Use the sidebar to select user and the voice input component above to record messages.")
-
-# ------------------ ArXiv Search Tab -------------------------
-with tabs[1]:
-    st.subheader("🔍 ArXiv Search")
-    query = st.text_input("Enter Query:")
-    if query and st.button("🔍 Search ArXiv"):
-        with st.spinner("Searching..."):
-            results = arxiv_search(query)
-            if results:
+            # 2. Perform ArXiv search automatically
+            with st.spinner("Searching ArXiv..."):
+                results = arxiv_search(voice_text)
                 summary = summarize_arxiv_results(results)
                 # Save as response
                 save_input_as_md(st.session_state['user_name'], summary, prefix="arxiv")
                 st.write(summary)
-                # Read aloud summary
+
+                # 3. Convert summary to audio and auto-play
                 voice = USER_VOICES.get(st.session_state['user_name'], "en-US-AriaNeural")
                 audio_file = speak_with_edge_tts(summary, voice=voice)
                 if audio_file:
                     play_and_download_audio(audio_file)
-            else:
-                st.warning("No results found.")
+
+            # 4. Update old_val to avoid repeated searches for same input
+            st.session_state['old_val'] = voice_text
+
+            # 5. Clear displayed text and re-run so next utterance starts fresh
+            st.experimental_rerun()
+
+    st.write("Speak a query to automatically run an ArXiv search and read results aloud.")
 
 # ------------------ History Tab -------------------------
-with tabs[2]:
+with tabs[1]:
     st.subheader("💾 History")
     files = list_saved_inputs()
     conversation = []
@@ -225,7 +203,6 @@ with tabs[2]:
         user, ts, content = parse_md_file(fpath)
         conversation.append((user, ts, content, fpath))
 
-    # Show conversation and read aloud each line
     for i, (user, ts, content, fpath) in enumerate(reversed(conversation), start=1):
         with st.expander(f"{ts} - {user}", expanded=False):
             st.write(content)
@@ -235,7 +212,6 @@ with tabs[2]:
                 if audio_file:
                     play_and_download_audio(audio_file)
 
-    # Read entire conversation
     if st.button("📜 Read Entire Conversation"):
         conversation_chrono = list(reversed(conversation))
         mp3_files = []
@@ -254,9 +230,9 @@ with tabs[2]:
             play_and_download_audio(combined_file)
 
 # ------------------ Settings Tab -------------------------
-with tabs[3]:
+with tabs[2]:
     st.subheader("⚙️ Settings")
-    st.write("Adjust parameters in the sidebar. Currently, no other settings to configure.")
+    st.write("Currently no additional settings. Use the sidebar to pick a user.")
 
 if st.session_state.should_rerun:
     st.session_state.should_rerun = False